Flaviviruses are members of the genus Flavivirus, which is classified within the family Flaviviridae. The flaviviruses are largely pathogenic to humans and other mammals. Flaviviruses that inflict disease on humans include yellow fever virus, JEV, dengue virus (including the four serotypes dengue-1, dengue-2, dengue-3 and dengue-4), tick-borne encephalitis virus, St. Louis encephalitis virus (SLEV), and others. Altogether there are about 70 species currently identified (Kuno et al., J. of Virol. 72: 73-83 (1998)).
The flaviviruses generally contain three structural proteins: M, the matrix or membrane protein; E, the envelope protein; and C, the capsid protein (Monath, in “Virology” (Fields, ed.), Raven Press, New York, 1990, pp. 763-814; Heinz and Roehrig, in “Immunochemistry of Viruses II: The Basis for Serodiagnosis and Vaccines” (van Regenmortel and Neurath, eds.), Elsevier, Amsterdam, 1990, pp. 289-305). M has a molecular weight (MW) of about 7-8 kDa; and E has a MW of about 55-60 kDa. M is synthesized as a larger precursor termed prM. The additional portion of prM is processed in the host cell to form M prior to secretion of mature virions. M and E are found in the membrane or envelope of the flavivirus particle, and so have long been considered to constitute important immunogenic components of the viruses.
The flaviviruses are RNA viruses whose single stranded RNA has a length, among the various species, of about 10 kb. The C protein, whose MW is 12-14 kDa, complexes with the RNA to form a nucleocapsid complex. Several nonstructural proteins are also encoded in the RNA genome; they are termed NS1, NS2A, NS2B, NS3, NS4A, NS4B and NS5. The genome is translated within the host cell as a polyprotein, then processed co- or post-translationally into the individual gene products by viral- or host-specific proteases (FIG. 1).
The nucleotide sequences of the genomes of several flaviviruses are known, as summarized in U.S. Pat. No. 5,494,671. That for JEV is provided by Sumiyoshi et al. (Virology 161:497-510 (1987)) and Hashimoto et al. (Virus Genes 1, 305-317 (1988)). The nucleotide sequences of the virulent strain SA-14 of JEV and the attenuated strain SA-14-14-2 used as a vaccine in the People's Republic of China are compared in the work of Nitayaphan et al. (Virology 177: 541-552 (1990)).
Nucleotide sequences encoding the structural proteins of other flavivirus species are also known. In many cases the sequences for the complete genomes have been reported. The sequences available include dengue type 1 virus (Mason et al., Virology 161:262-267 (1987)), dengue type 2 virus (Deubel et al., Virology 155:365-377 (1986); Gruenberg et al. J. Gen. Virol. 69, 1391-1398 (1988); Hahn et al. Virology 162, 167-180 (1988)), dengue type 3 virus (Osatomi et al., Virus Genes 2:99-108 (1988)), dengue type 4 virus (Mackow et al., Virology 159:217-228 (1987); Zhao et al. Virology 155, 77-88 (1986)), and yellow fever virus (YFV) (Rice et al., Science 229:726-733 (1985)).
Many flaviviruses including JEV are transmitted to humans and other host animals by mosquitoes. They therefore occur over widespread areas, and their transmission is not easily interrupted or prevented. JEV affects adults and children, and there is a high mortality rate among infants, children, and the elderly, in areas of tropical and subtropical Asia (Tsai et al., in “Vaccines” (Plotkin, ed.) W. B. Saunders, Philadelphia, Pa., 1994, pp. 671-713). Among survivors, there are serious neurological consequences, related to the symptoms of encephalitis, that persist after infection. In more developed countries of this region such as Japan, the Republic of China, and Korea, JEV has been largely controlled by use of a vaccine of inactivated JEV. Nevertheless, it is still prevalent in other countries of the region.
Dengue virus disease is also mosquito-borne, occurring globally in regions with tropical and sub-tropical climates. Symptoms include fever, rash, severe headache and joint pain, but mortality from dengue is low. Epidemics of dengue virus are sufficiently frequent and widespread that the disease represents a major public health problem. Nevertheless, safe and effective vaccines to protect against dengue are not available, despite decades of effort. There thus is a strong need for a vaccine against dengue.
Yellow fever is prevalent in tropical regions of South America and sub-Saharan Africa, and is transmitted by mosquitoes. Infection leads to fever, chills, severe headache and other pains, anorexia, nausea and vomiting, with the emergence of jaundice. A live virus vaccine, 17D, grown in infected chicken embryos, is considered safe and effective. Nevertheless, there remains a need for a vaccine that avoids the necessity of administering live virus, with its attendant development of mild symptoms and viremia.
The vaccines available for use against JEV include live virus inactivated by such methods as formalin treatment as well as attenuated virus (Tsai et al.). Whole virus vaccines, although effective, do have certain problems and/or disadvantages. The viruses are cultivated in mouse brain or in cell culture using mammalian cells as the host. Such culture methods are cumbersome and expensive. Furthermore, there is the attendant risk of incorporating antigens from the host cells, i.e., the brain or other host, into the final vaccine product, potentially leading to unintended and undesired allergic responses in the vaccine recipients. There is also the risk of inadvertent infection among workers involved in vaccine production. Finally, there is the risk that the virus may not be fully or completely inactivated or attenuated, and thus, the vaccine may actually cause disease.
A recombinant flavivirus which is a chimera between two flaviviruses is disclosed in WO 93/06214. The chimera is a construct fusing non-structural proteins from one “type”, or serotype, of dengue viruses or a flavivirus, with structural proteins from a different “type”, or serotype, of dengue virus or another flavivirus. The second flavivirus may be JEV.
Several recombinant subunit and viral vaccines have been devised in recent years. U.S. Pat. No. 4,810,492 describes the production of the E glycoprotein of JEV for use as the antigen in a vaccine. The corresponding DNA is cloned into an expression system in order to express the antigen protein in a suitable host cell such as E. coli, yeast, or a higher organism cell culture. U.S. Pat. No. 5,229,293 discloses recombinant baculovirus harboring the gene for JEV E protein. The virus is used to infect insect cells in culture such that the E protein is produced and recovered for use as a vaccine.
U.S. Pat. No. 5,021,347 discloses a recombinant vaccinia virus into whose genome the gene for JEV E protein has been incorporated. The live recombinant vaccinia virus is used as the vaccine to immunize against JEV. Recombinant vaccinia and baculoviruses in which the viruses incorporate a gene for a C-terminal truncation of the E protein of dengue type 2, dengue type 4, and JEV are disclosed in U.S. Pat. No. 5,494,671. U.S. Pat. No. 5,514,375 discloses various recombinant vaccinia viruses which express portions of the JEV open reading frame extending from prM to NS2B. These pox viruses induced formation of extracellular particles that contain the processed M protein and the E protein. Two recombinants encoding these JEV proteins produced high titers of neutralizing and hemagglutinin-inhibiting antibodies, and protective immunity, in mice. The extent of these effects was greater after two immunization treatments than after only one. Recombinant vaccinia virus containing genes for the M and E proteins of JEV conferred protective immunity when administered to mice (Konishi et al., Virology 180: 401-410 (1991)). HeLa cells infected with recombinant vaccinia virus bearing genes for prM and E from JEV were shown to produce subviral particles (Konishi et al., Virology 188: 714-720 (1992)). Dmitriev et al. report immunization of mice with a recombinant vaccinia virus encoding structural and certain nonstructural proteins from tick-borne encephalitis virus (J. Biotechnol. 44:97-103 (1996)).
Recombinant virus vectors have also been prepared to serve as virus vaccines for dengue fever. Zhao et al. (J. Virol. 61, 4019-4022 (1987)) prepared recombinant vaccinia virus bearing structural proteins and NS1 from dengue type 4 and achieved expression after infecting mammalian cells with the recombinant. Similar expression was obtained using recombinant baculovirus infecting target insect cells (Zhang et al. J. Virol. 62, 3027-3031 (1988)). Bray et al. (J. Virol. 63, 2853-2856 (1989)) also report a recombinant vaccinia dengue vaccine based on the E protein gene that confers protective immunity on mice when challenged to develop dengue encephalitis. Falgout et al. (J. Virol 63, 1852-1860 (1989)) and Falgout et al. J. Virol. 64, 4356-4363 (1990) report similar results. Zhang et al. (J. Virol 62, 3027-3031 (1988)) showed that recombinant baculovirus encoding dengue E and NS1 proteins likewise can protect mice against dengue encephalitis when challenged. Other combinations in which structural and nonstructural genes are incorporated into recombinant virus vaccines fail to produce significant immunity (Bray et al. J. Virol. 63, 2853-2856 (1989)). Also, monkeys failed to develop fully protective immunity to dengue virus challenge when immunized with recombinant baculovirus expressing the E protein (Lai et al. (1990) pp. 119-124 in F. Brown, R. M. Chancock, H. S. Ginsberg and R. Lerner (eds.) “Vaccines 90: Modern approaches to new vaccines including prevention of AIDS”, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.).
Immunization using recombinant DNA preparations has been reported for St. Louis encephalitis virus (SLEV) and dengue-2 virus using weanling mice as the model (Phillpotts et al., Arch. Virol. 141: 743-749 (1996); Kochel et al., Vaccine 15: 547-552 (1997)). Plasmid DNA encoding the prM and E genes of SLEV provided partial protection against SLEV challenge with a single or double dose of DNA immunization. In these experiments control mice exhibited about 25% survival, and no protective antibody was detected in the DNA immunized mice (Phillpotts et al.). In mice that received three intradermal injections of recombinant dengue-2 plasmid DNA containing prM, 100% developed anti-dengue-2 neutralizing antibodies, and 92% of those receiving the corresponding E gene likewise developed neutralizing antibodies (Kochel et al.). Challenge experiments using a two-dose schedule, however, failed to protect mice against lethal dengue-2 virus challenge.
The vaccines developed to date for immunizing against JEV have a number of disadvantages and problems attending their use. Inactivated virus vaccine is costly and inconvenient to prepare. In addition, it carries the risk of allergic reaction originating from proteins of the host used in preparing the virus. Furthermore, it presents considerable risk to the workers employed in their production. Candidate attenuated JEV vaccines are undergoing clinical trials but as of 1996 have not found wide acceptance outside of the People's Republic of China (Hennessy et al., Lancet 347: 1583-1586 (1996)). Recombinant vaccines based on the biosynthetic expression of only certain of the proteins of the JEV genome appear not to induce high antibody titers, and, as with the whole virus preparations, carry the risk of adverse allergic reaction to antigens from the host organism, or to the vaccinia virus vector, as the case may be. Similar problems attend the preparation of vaccines against YFV. Vaccine development against dengue is less advanced, and such virus-based or recombinant protein-based vaccines face similar problems as those just alluded to.
There is therefore a need for vaccines directed against flaviviruses such as yellow fever, dengue, JEV, and SLEV which are inexpensive to prepare, present little risk to workers involved in their manufacture, carry minimal risk of adverse immunological reactions due to impurities or adventitious immunogenic components, and are highly effective in eliciting neutralizing antibodies and protective immunity. There is furthermore a need for a vaccine against JEV and related flaviviruses that minimizes the number of immunizing doses required.